Parallel Multi-Label Propagation Based on Influence Model and Its Application to Overlapping Community Discovery

2017 ◽  
Vol 26 (03) ◽  
pp. 1760013 ◽  
Author(s):  
Qirong Qiu ◽  
Wenzhong Guo ◽  
Yuzhong Chen ◽  
Kun Guo ◽  
Rongrong Li
Keyword(s):  
Complex Network ◽  
Label Propagation ◽  
Community Discovery ◽  
Influence Model ◽  
Propagation Algorithm ◽  
Overlapping Community ◽  
High Scalability ◽  
Computationally Intensive ◽  
Mapreduce Model ◽  
Intensive Work

Finding communities in networks is one of the challenging issues in complex network research. We have to deal with very large networks that contain billions of vertices, which makes community discovery a computationally intensive work. Moreover, communities usually overlap each other, which greatly increases the difficulty of identifying the boundaries of communities. In this paper, we propose a parallel multi-label propagation algorithm (PMLPA) that enhances traditional multi-label propagation algorithm (MLPA) in two ways. First, the critical steps of MLPA are parallelized based on the MapReduce model to get higher scalability. Second, new label updating strategy is used to automatically determine the most valuable labels of each vertex. Furthermore, we study the improvement of PMLPA through considering the influence of vertices and labels on label updating. In this way, the importance of each label can be described with higher precision. Experiments on artificial and real networks prove that the proposed algorithms can achieve both high discovering accuracy and high scalability.

Community leader and transition probability based LPA

2020 ◽  
Vol 34 (27) ◽  
pp. 2050253
Author(s):  
Yu Ying Chen ◽  
Jimin Ye
Keyword(s):  
Transition Probability ◽  
Real Data ◽  
Label Propagation ◽  
Community Leader ◽  
Data Sets ◽  
Community Discovery ◽  
Normalized Mutual Information ◽  
Network Community ◽  
Propagation Algorithm ◽  
Close Relationship

Many practice problems can be transformed into complex networks, and complex network community discovery has become a hot research topic in various fields. The classic label propagation algorithm (LPA) can give community partition very quickly, but stability of the algorithm is poor due to random label propagation. To solve this problem, community leader principle is built and transition probability is introduced, a label propagation algorithm based on community leader and transition probability (CTLPA) is proposed. CTLPA selects threatened leaders and their communities according to the community leader principle, and uses the transition probability and the degree of the leader to jointly control the order for community merger, so that the threatened leader continuously devours the communities that threaten him, until a preliminary community partition is formed. To further reduce the number of community, in CTLPA, based on the characteristic of the community structure: close relationship within the community and sparse relationship outside the community, the closest communities are merged, until the final community partition is obtained. The CTLPA is compared with other five classic algorithms on LFR artificially generated networks and several real data sets. The experimental results show that CTLPA is robust in community partition, it always gives the same community partition, while the LPA will give different results from multiple independent runs. The number of community partition and the normalized mutual information (NMI) of the CTLPA are the best in most cases.

scholarly journals Overlapping Community Detection Based on Membership Degree Propagation

Entropy ◽  
2020 ◽  
Vol 23 (1) ◽  
pp. 15
Author(s):  
Rui Gao ◽  
Shoufeng Li ◽  
Xiaohu Shi ◽  
Yanchun Liang ◽  
Dong Xu
Keyword(s):  
Complex Network ◽  
Community Detection ◽  
Real World ◽  
Detection Algorithm ◽  
Label Propagation ◽  
Computational Time ◽  
Membership Degree ◽  
Overlapping Community Detection ◽  
Overlapping Community ◽  
Real World Datasets

A community in a complex network refers to a group of nodes that are densely connected internally but with only sparse connections to the outside. Overlapping community structures are ubiquitous in real-world networks, where each node belongs to at least one community. Therefore, overlapping community detection is an important topic in complex network research. This paper proposes an overlapping community detection algorithm based on membership degree propagation that is driven by both global and local information of the node community. In the method, we introduce a concept of membership degree, which not only stores the label information, but also the degrees of the node belonging to the labels. Then the conventional label propagation process could be extended to membership degree propagation, with the results mapped directly to the overlapping community division. Therefore, it obtains the partition result and overlapping node identification simultaneously and greatly reduces the computational time. The proposed algorithm was applied to a synthetic Lancichinetti–Fortunato–Radicchi (LFR) dataset and nine real-world datasets and compared with other up-to-date algorithms. The experimental results show that our proposed algorithm is effective and outperforms the comparison methods on most datasets. Our proposed method significantly improved the accuracy and speed of the overlapping node prediction. It can also substantially alleviate the computational complexity of community structure detection in general.

LinkLPA: A Link-Based Label Propagation Algorithm for Overlapping Community Detection in Networks

2016 ◽  
Vol 33 (2) ◽  
pp. 308-331 ◽  
Author(s):  
Heli Sun ◽  
Jiao Liu ◽  
Jianbin Huang ◽  
Guangtao Wang ◽  
Xiaolin Jia ◽  
...  
Keyword(s):  
Community Detection ◽  
Label Propagation ◽  
Overlapping Community Detection ◽  
Propagation Algorithm ◽  
Overlapping Community
Sign in / Sign up

Export Citation Format

Share Document